Contingencies of Value Devices and Conventions at a Design School Admission Test

Based on a study of the admission test at a design school, this paper investigates the contingencies of aesthetic values as these become visible in assessment practices. Theoretically, the paper takes its starting point in Herrnstein Smith’s notion of ‘contingencies of values’ and outlines a pragmatist ground where cultural sociology and economic sociology meet. Informed by the literature on cultural intermediaries, the paper discusses the role of evaluators and the devices which accompany them. Whereas studies of cultural intermediaries traditionally apply a Bourdieusian perspective, recent developments within this fi eld of literature draws inspiration from the so-called ‘new new economic sociology,’ which this paper adds to. While the admission test is easily described as a matter of overcoming “subjective” aesthetic evaluations by means of “objective” and standardized assessment criteria, the paper does not accept this storyline. As an alternative, the paper outlines the contingencies of values which are at play at the admission test, composed of of fi cial assessment criteria and scoring devices together with conventions within the world of design, and set in motion by interactions with the objects that applicants submit. The new petite bourgeoisie comes into its own in all the occupations involving presentation and representation (sales, marketing, advertising, public relations, fashion, decoration and so forth) and in all the institutions providing symbolic goods and services . . . and in cultural production and organization which have expanded considerably in recent years. (1984, 359)

artists valuations of their work compose important moments in the transition from hopeful aspiration to manifestations of an artistic career. In that way, the act of valuation consists in considering a reality while provoking it (Muniesa 2012, 32; see also Michael 2000), as valuators' decisions play a part in composing the future for hopeful candidates.
Admission tests at art and design schools exemplify one of the places where aesthetic valuations take place and become observable. However, quite little is known about how assessment criteria at art and design schools are developed and applied (Harland and Sawdon 2011). To consider these issues about the formation and employment of assessment criteria the paper looks into a case of the admission test at a prominent design school in Denmark. To enter the school applicants go through a two-tier test. During the first round of the test applicants make a home assignment on a set subject where they remain anonymous. During the second round of the test a number of applicants are invited to interviews based on their results from the first round. Over the last years, the school has worked at standardizing and explicating its assessment criteria to make them transparent and nonsubjective. Thus, the paper looks into the scoring mechanisms that the school is elaborating on and investigates how assessments are carried out in practice.
To conceptualize the role of evaluators at the admission test, the paper employs the notion of cultural intermediaries. Cultural intermediaries mediate between production and consumption, and the term thus includes a broad and diverse group of professions involved in qualifying cultural goods (Matthews and Maguire 2012; Nixon and du Gay 2002). A subsection of cultural intermediaries are characterized by their actions as decision-makers, and evaluators fall under this category. As the term 'cultural intermediaries' stems from Bourdieu (1984), there is a well-established tradition for analyzing the role that cultural intermediaries play with a Bourdieusian framework highlighting habitus. Some of the empirical material from the admission test at the design school can be interpreted in a way that fits this framework very well. However, there seems to be more to the story than this as the school actively strives to professionalize and standardize its evaluations and evaluators operate with a clear set of conventions.
To open up and reconsider the Bourdieusian definition of cultural intermediaries, the paper draws inspiration from developments of the concept informed by the 'new new economic sociology' inspired by actor-network theory (ANT) (McFall 2009; see also Callon et al. 2002;Cronin 2004;Moor 2012). Based on this new approach studies have been focusing on the plurality of devices which cultural intermediaries employ, turning attention from dispositions to devices (du Gay 2004). In line with this approach, the paper considers the design school's attempt to standardize its evaluations by means of three devices: a list of official assessment criteria and two scoring mechanisms based on these criteria. However, when the paper inquires how these devices are employed in practice, it turns out that assessments do neither start from nor center on the official criteria. Sporadically, evaluators emphasize an official criterion during discussions. Yet, most frequently scoring mechanisms are not introduced until after assessments have been made. That means evaluators do not found their assessments on the official criteria but rather translate their assessments into the official criteria subsequently. When making assessments evaluators rely on a set of parameters that can be seen as conventions in the design world (Becker 1982). That is, as evaluators are met with piles of submissions they employ a repertoire of practical and operational valuation principles based in their professional expertise. The paper outlines this set of parameters that evaluators use in making assessments, discussing how they are translated into official criteria and measurable entities.
To consider this situation of mobilizing several assessment criteria in the forms of both devices and conventions the paper subscribes to a pragmatist approach. Pragmatism constitutes a theoretical foundation where economic sociology and cultural sociology meet, as pragmatist scholars within both sub-disciplines have suggested seeing valuations as contingent (Herrnstein Smith 1988;Muniesa 2012;. The pragmatist approach suggests that objects of valuations can act in many ways, yet these ways are traceable and based on affordances of the objects. This means that in contrast to the Bourdieusian tradition where aesthetic valuations are seen as deriving from habitus, which makes the objects secondary and risks portraying assessments as subjective, the pragmatist approach turns attention to the specific contingencies that form values.
Structurally, the first section of the paper explicates the theoretical underpinnings of the analysis. Following after that, the admission test and the study which the paper builds on are presented. Subsequently, the analysis falls in three parts: The first part presents a Bourdieusian interpretation of assessments made during the test which is in line with the traditional view on cultural intermediaries. Yet, this section also introduces the evaluators' objections to the Bourdieusian analysis of their work, thus indicating the need for alternative approaches. Accordingly, the second part considers the devices of standardized assessment criteria and two scoring mechanisms which the school introduces to overcome personal bias. This section follows the 'new new economic sociology' as an alternative approach to studying the work of cultural intermediaries. Yet, in the third part, the analysis outlines the set of operational parameters which evaluators rely on when making assessments. These parameters represent conventions within the design world, rather than criteria implemented by the school. To conclude with, the paper suggests seeing assessments made during the admission test as an exercise in contingencies of values.
Theoretical Inspirations: Cultural Inter mediar ies and Quantif ication of Aest hetic Qualities Rather than seeing aesthetic value as something that belongs only to its own sphere, governed by its own set of rules, Herrnstein Smith suggests that aesthetic valuations pertain to valuations in general (1988). Thus, investigations of aesthetic evaluations may contribute to a general rethinking of the concept of value (ibid., 28). As an alternative to seeing aesthetic valuations as pure, non-utilitarian, and interest-free, deriving from intellectual, sensory or perceptual activities, Herrnstein Smith proposes tracing the continuity and stability, as well as shifts and diversities, of aesthetic values (ibid., 33ff.). To pursue this alternative route, Herrnstein Smith calls for studies that investigate the contingencies of value: If we realize that literary value is 'relative' in the sense of contingent (that is, a changing function of multiple variables) rather than subjective (that is, personally whimsical, locked into the consciousness of individual subjects and/or without interest or value for other people), then we may begin to investigate the dynamics of that relativity. (1988, 11, emphasis in the original) Following Herrnstein Smith's approach, this paper illustrates how a design school aims to transcend subjective evaluations by introducing standardized assessment criteria. To consider this situation, the notion of devices from economic sociology is useful as it calls attention to mechanisms that aim to stabilize valuations ). In the case of the admission test, the introduction of a list of official assessment criteria accompanied by two scoring mechanisms can be seen exactly as an attempt to stabilize valuations by means of these devices. However, the paper suggests that it is not a simple matter to employ these devices in assessment practices. The devices which the school has introduced do not structure assessments in-the-making. Rather, evaluators use the devices of the official assessment criteria and scoring mechanisms after making their assessments. While being in the process of assessing, evaluators focus on the work in question, considering its specifics. To assess these specifics evaluators rely on a set of parameters that can be seen as conventions within the world of design (Becker 1982). Although these conventions are not written down they constitute a shared repertoire that is used by all evaluators during the admission test. Accordingly, the paper suggests that the contingencies of values at the admission test depend not on the devices which the school introduces but also the conventions that evaluators rely on.
Pragmatism constitutes the theoretical foundation for this paper and its agenda of looking into contingencies of values. Moreover, this theoretical foundation presents a shared ground between cultural sociology and economic sociology. In recent developments economic sociology has turned to pragmatism, in particular Dewey, to outline valuation as an action (Muniesa 2012). Conventionally, a binary opposition characterizes studies of value, as value is considered either as something that something has as a result of its own condition or as something that something has by virtue of how people consider it (ibid., 24). As an alternative, the pragmatist approach suggests seeing valuation as a form of mediation, a process of doing something to something else (ibid., 32). Seeing qualities of products as both intrinsic and extrinsic (Callon et al. 2002), the pragmatist approach thus suggests investigating actions of valuations where the value of something is considered and at the same time provoked (Muniesa 2012). In parallel to this new pragmatist approach within economic sociology, cultural sociology has undergone a material turn in recent years based on a growing interest in the active role of cultural products, which has produced a new, post-Bourdieusian approach (Born 2010), a strand of which may be described as pragmatist (Hennion 2004). Whereas cultural products have often been described either as a stimulus capable of working independently of its circumstances or as a result of social causes and thus a transmitter powers and meaning, the new cultural sociology suggests that cultural objects are at the same time constructed objects and generating various effects (DeNora 2000;Hennion and Grenier 2000). Thus, similar to the new pragmatist approach within economic sociology, the new pragmatist approach within cultural sociology proposes to move beyond seeing value either as inherent in the thing itself or constructed by users. Rather, the pragmatist approach sets out to investigate the co-production of cultural products and their users (Hennion 2001), looking into how cultural products enable actions and simultaneously become constructed by these actions. While Herrnstein Smith's book on aesthetic value Contingencies of Value (1988) has not formed part of the new pragmatist strand within cultural sociology, it may provide a useful furtherance of this strand of research. Arguing that aesthetic value is neither solely intrinsic nor extrinsic but a function of contingent relations, Herrnstein Smith's pragmatist approach, inspired by Dewey, corresponds well with the pragmatist strands within both economic sociology and cultural sociology.

Cultural Intermediaries
As cultural sociologists have described with notions such as the decision chain model (Ryan and Peterson 1982), series of mediations (Hennion 1997), and regimes of mediation (Cronin 2004), cultural products undergo a number of alterations during production and distribution-and various cultural intermediaries take part in defining these alterations. At each stage in the processes of creation and dissemination do new cultural intermediaries encounter and form cultural products.
Compared to a Latourian terminology, the concept of cultural intermediaries may cause confusion as Latour distinguishes mediators from intermediaries, suggesting that intermediaries do not transform that which they transport (2005, 37ff.). Yet, cultural intermediaries function precisely as mediators, connecting while altering that which they connect. In opposition to traditional gatekeepers, who can be seen as intermediaries in the Latourian sense, Negus suggests that cultural intermediaries become involved when cultural products are still unfinished and thus play a part in shaping products (2002a).
However, the admission test can be described as gatekeeping in the most traditional sense: choosing between products that candidates have aimed to finish and regulating access by being in power to decide who is in and who is out (DiMaggio and Hirsch 1976; see also Crane 1992;Peterson 1994). Whereas the evaluators work as lecturers at the design school and illustrate one type of cultural intermediaries during the school year, when they initiate, guide and supervise the work of students, their role changes into gatekeeping at the admission test when they select products and candidates. Here, the lecturers' primary task is to decide which applicants to accept; their valorizations are means to reach this end, and for that reason they are spoken of as evaluators during the test. In this capacity, the lecturers illustrate a specific type of intermediaries whose core activity consists in producing evaluations, rankings and selections (Bessy and Chauvin 2013, 101).
As they assess and select candidates, evaluators are in a powerful position that can be compared to other decision-making cultural intermediaries. As Beckert and Aspers suggest: Aesthetic markets (Aspers 2001, 1) generally have no objective standards by which quality could be measured and compared. Instead, quality is constructed from the judgments of the participating actors. As gallery owners, museum curators, art critics, collectors, or professors at art schools, these experts shape the evaluation on art works through their opinions, reviews, purchasing decisions, and exhibition policies. These authorities are carefully watched by the other actors in the field, who deduce the quality of an artist through the judgments of professionals. (2011,20) Like critics, whose expert-opinions may become self-fulfilling prophecies; evaluators verbalize their judgments and invent criteria that decide the fate of candidates (Hutters 2011;Karpik 2010). Also similar to high-status editors in the world of fashion, whose promotion of chosen models and photographers is decisive, evaluators assign value to and stabilize value of candidates (Aspers 2001; Mears 2011).

The Bourdieusian Approach to Cultural Intermediaries
The term cultural intermediary originates from Bourdieu and was presented in his legendary book Distinction to characterize a growing group of professions, a new petite bourgeoisie, covering a broad range of occupations involved in providing symbolic goods and services: The new petite bourgeoisie comes into its own in all the occupations involving presentation and representation (sales, marketing, advertising, public relations, fashion, decoration and so forth) and in all the institutions providing symbolic goods and services . . . and in cultural production and organization which have expanded considerably in recent years. (1984,359) A characteristic feature of this new class fraction is a relative openness in terms of entry qualifications. Blurring the line between high art and popular culture, and bridging the gap between personal taste and professional judgment (Negus 2002a, 503) these new occupations signal heterogeneity. As Nixon explains it: For Bourdieu, then, these occupations stand out in being composed of a mixture of déclassé middle-class individuals and socially aspirational individuals from lower middle-class and, particularly, working-class backgrounds. It is this social mix that gives these occupations much of their distinctive character. (Nixon 2003, 60) Based on their stirring of social stratification, jobs in media and cultural industries "are popularly regarded as cool, creative and egalitarian" (Gill 2002, 70). As iconic manifestations of a fuse between leisure and work, cultural work has come to signal a new creative era (Featherstone 1991;Florida 2002). Yet, as an alternative to formal entry qualifications, recruitment to these jobs depends on network which makes social and cultural capital decisive (Lee 2013). Hence, despite their cool and egalitarian image, the flexibility and informality of these occupations reproduce inequalities (Gill 2002). As studies of the advertising industry suggest, a "taken-for-granted middle class nature" subsists (Cronin 2004, 353;Nixon 2003).
To empirically ground the claims about the role of cultural intermediaries a number of studies have investigated how cultural intermediaries carry out their jobs. For example, Negus demonstrates, in a study of music production companies in the UK and the USA, how studio executives make choices about which artists to promote depending on their own cultural dispositions. As the senior executives are predominantly white, middle-aged, middle-class, males who were young in the 60s when rock bands gained ground, they choose rock bands at the expense of pop, soul and r'n'b artists, despite the preferences of the market in the 90s (2002a; 2002b). In that way, habitus is confirmed as the most important factor behind aesthetic valuations.
Characteristically, the Bourdieusian perspective proposes a critical approach. In this perspective, taste is not neutral but a central feature of the power struggle to occupy constitutive positions within a field (Bourdieu 1993). Cultural capital in the Bourdieusian sense means familiarity with the dominant culture in society and, similar to other educational institutions, art and design schools may reproduce social differences by assuming the possession of cultural capital (Zimdars et al. 2009). Thus, evaluators' preferences as well as candidates' abilities to identify and fulfill these at admission tests may be interpreted as a matter of cultural capital.

A New Approach to Cultural Intermediaries Inspired by the 'New New Economic Sociology'
A relatively new approach to cultural intermediaries has been developed inspired by what McFall calls the 'new new economic sociology' informed by ANT (McFall 2009). Although relatively few studies have pursued this alternative route (Cronin 2004 andMoor 2012 are important examples), it has been described as promising for revitalizing and progressing the concept of cultural intermediaries (du Gay 2004;Matthews and Maguire 2012). A central question within the 'new new economic sociology' is how market designs influence processes of qualification (Cochoy 2008;MacKenzie 2009, Muniesa 2007. When this question is transported to studies of cultural intermediaries it widens the scope of study, suggesting that several other factors than cultural intermediaries themselves may be relevant to address to understand their actions and influence. As Cronin describes it (2004), the 'new new economic sociology' has been studying how market actors are involved in the qualification of products (Callon et al. 2002), which parallels the discussion about how cultural intermediaries carry out mediations. Yet, in contrast to studies of cultural intermediaries, which have focused on the habitus of cultural intermediaries, studies within the 'new new economic sociology' have been highlighting the active role of devices such as measurement techniques, pricing models and merchandising tools (Muniesa et al. 2007, 2). Inspired by this, attention has been turned to devices that accompany cultural intermediaries (see for example Moor 2012).
Adding to this new perspective on cultural intermediaries, the analysis in this paper considers the quantification of qualitative features of aesthetic products. By investigating the official assessment criteria that the school operates with, and the two ranking systems that accompany it, the analysis looks into the school's taxonomy of qualities and how it is employed as a calculative device to rank applicants. Investigating the translation from aesthetic products to scores, the paper illustrates the production of data; how candidates are produced by the evaluators' assessment practices assisted by the official assessment criteria and ranking systems that the school has invented (Ruppert 2011). In that way, the paper exemplifies what a ranking system does by studying how aesthetic products, that are meant to be unique and singular, are equated and compared at the test (Didier 2010;Karpik 2010). Based on this, that analysis raises the question about how to classify and count aesthetic qualities (Desrosières 1998).

Introducing t he S tudy: The Case of t he Admission Test
A study of the admission test composes the basis for the paper. I conducted this study in 2010. Access to the admission test was given to me by the school's head of study program. I followed the admission test primarily by means of non-participant observations. Observations were carried out throughout a week in April when evaluators reviewed submitted folders and over three days in May when evaluators interviewed selected applicants. Evaluators were split into teams that worked alongside each other, and I followed a different team of evaluators each day. This means that the study entails only a part of the assessments that were made during the test, but nevertheless assessments performed by a variety of evaluators. During my observations I noted down exchange of words between evaluators alongside brief descriptions of the works that evaluators were looking at. As I had not gained permission from all applicants I was not allowed to photograph the submitted folders although it would have been a highly useful method of documentation. In addition to my observations, I talked to evaluators in breaks when they asked me about my study and I asked them to clarify things for me. Moreover, in June, I presented my observations to the lecturers and administration at the school at their internal evaluation of the test.
Evaluators gave me access to a confidential situation where they make numerous decisions in a short time, often stated in quite few and frank words. Similar to discussions of grades after examinations at other educational institutions, evaluations at the admission test did not dance around the issue but were to the point. This means that transcriptions of the evaluations when taken out of their original setting may sound harsher than when spoken in the moment of making the decision. Transcripts of specific evaluations constitute a particular type of empirical material which is quite different from, for example, interviews with evaluators about their practices in general. For this reason the name of the school has been anonymized.

About the Test
Every year before March fifteen candidates can register for the admission test via the Coordinated Enrolment System, which is used for accessing all higher education institutions in Denmark. Admittance into all higher education programs, including programs at the design school in question, requires a General Certificate of Secondary Education. If candidates meet this requirement, they receive an applicant number from the Coordinated Enrolment System, and this number is used to track applicants throughout the admission test.
At the beginning of April, the admission test is launched on the school webpage, and candidates have two weeks to produce their answers. The school makes a new assignment each year. Submissions should be between fifteen and twenty-five pages. Submissions can be handed in analogue or digital versions. The format of analogue submissions is optional but maximum size A3. The format of digital submissions is pdf and a maximum of twenty-five megabytes. Together with the response to the set assignment applicants can include up to three examples of their previous work. A full submission is referred to as a folder because submissions are traditionally delivered bound in a folder.
One week is set aside for going through all the submissions at the school. This takes place in April, right after the deadline for submissions. About ten lecturers participate each day, and they are split into teams of two. The teams are formed so that they consist of lecturers from different fields such as fashion design, interaction design and industrial design. Moreover, current students at the school participate as observers of the assessments. During the week in 2010 when I followed the test there were about as many students as lecturers participating on the first day and they contributed enthusiastically to the discussions. However, on the morning on the second day the school administration stressed that the students' role was only to observe assessments. After this reminder the participation of students declined; there were fewer students present at the assessments and they only took part in the discussions sporadically.
Generally, evaluators spend between five and fifteen minutes on each folder. How much time they spend depends on the quality of the folder. If the submission is judged to be very poor the evaluation is often done in a couple of minutes. Conversely, if the submission is considered competent and thorough, or difficult to assess, the assessment can take up to fifteen minutes. Most of the evaluations that I observed took a little less than ten minutes. On the first day only one hundred and twenty folders were evaluated, meaning that the evaluators had to speed up, assessing around two hundred folders each day for the next four days.
In May, selected applicants are invited to interviews. Similar to the process of reviewing folders, the interviews take place over a week with lecturers working as evaluators in teams of two. At the interview sessions, evaluators do not know how folders have been evaluated beforehand, and it is only by coincidence if an evaluator meets the same folder during the first and second round. Interviews take approximately twenty minutes. First, applicants present their submissions. Afterwards, evaluators inquire about previous schooling and work experiences, interests, and future plans, based on a set of standardized questions. After the interview, evaluators grade the applicant. The grading at the interview makes the final score at the test.
In 2010, 1257 candidates signed up for the admission test via the Coordinated Enrolment System, 703 handed in answers to the home assignment, 199 were selected for interviews, 150 applicants passed the test at the interview, and 105 were chosen to start at the school.

Cultural Capital S tands t he Test : The Bourdieusian Approac h
Without doubt, a number of empirical examples from the admission test suit a Bourdieusian analysis about social exclusion perfectly. Lecturers at the design school can be observed choosing applicants who possess the required cultural capital and excluding those who do not. Showing how evaluators' choices may support the Bourdieusian argument that cultural capital is decisive, two examples from the interview sessions will serve as illustrations.

Assessments of Two Interviewees
A chubby Greenlandic guy comes in. He has a crew cut, wears a hoodie and speaks with a Jutlandic dialect. He explains that he has had a brief education in design from a technical school and has been self-employed in advertising for some years, but currently he is employed at a convenience store and would like to return to school to develop his skills. His assignment consists in clothing equipment for expat aid workers in conflict and disaster zones. The clothing is meant to be comfortable, warm and at the same time signal "authority in a non-threatening way," the applicant explains. At the evaluation, after a remarkably short interview, the first evaluator begins: "It left me speechless. He is good enough, but kind of stiff." The second evaluator adds: "Really shy also, right. I'll write 'nerd'." The first evaluator returns to the assignment: "It's supposed to be non-frightening and then comes Terminator [a heavy piece of clothing that the evaluators associate with science-fiction and militarism]." They each give their grades and the candidate receives 17 points in total. The first evaluator concludes: "17then he won't get in. Last year the [cutoff] line went at 18." Two days later, when I follow a different team of evaluators, a tall, lean guy with slicked back hair comes in. He is wearing laced boots and a white shirt with a beige V-Neck sweater. He explains that he has been attending the Scandinavian Design College over the past half year, after finishing high school. Asked which subject he is interested in, he answers: "Industrial design, I have told myself. My mum also works for a graphic design company." His assignment consists in a re-design of a pavilion for the Roskilde Festival made of biodegradable materials. He also shows a couple of assignments he has produced at the design college. Asked what he is inspired by, he answers: "I've just seen the exhibition on green architecture at Louisiana [Louisiana Museum of Modern Art, which is the most visited art museum in Denmark]." The evaluator follows up: "Are you interested in that [green architecture]?" The applicant responds: "Louisiana is SUCH a great place; I just cannot go there often enough! Sustainability interests me, but it can also get too much." After the interview the applicant begins rolling down the sleeves of his shirt, buttoning them and rolling down the sleeves of his sweater over the sleeves of the shirt. He takes his time, in contrast to the other applicants who have hurried out. At the evaluation the first evaluator starts: "He's a likely student . . . but lacks creativity . . . It's also the milieu he comes from, he's in the 7's [middle region]." The second evaluator agrees: "Definitely passed, he has something to offer, he's in the top section." The first evaluator hesitates: "He also had weak points, but he's used to talking." The second evaluator agrees: "It's not in the detail, not aesthetic considerations." The first evaluator decides: "But it's fascinating with someone who understands the context." The candidate gets 7 four times, which gives a total of 28 points. The first evaluator concludes: "Then it'll go right through." What happens during these two instances of evaluation? According to the evaluators, the chubby applicant is "good enough," but obviously that is not sufficient to grant him access to the school. Technical qualifications become overruled by an estimation of the applicant's personality, which is deemed nerdy and shy. After they have reached their decision, I ask the evaluators whether they select students that fit into the school socially: if blending into the body of students at the school is a criterion. The first evaluator replies: "He would have a really hard time here." The second evaluator agrees.
On the other hand, the evaluators estimate that the applicant with the slicked back hair "lacks creativity," but he is accepted into the school nevertheless. Despite his "weak points" and lack of "aesthetic considerations," he is "used to talking" and able to put things in "context" and that is apparently more important. As one of the evaluators point out, this applicant comes from a "milieu" that gives him an advantage: with a parent working in the industry, frequent visits to the most famous art museum in the country and having completed a course at one of the best pre-schools you get acquainted with the world of the school. In sharp contrast with the chubby applicant, the applicant with the slicked back hair demonstrates a great amount of confidence that is based solidly on his literacy of the field.
In these examples, assignments are interpreted as signs of whether the applicant is familiar with the design field. The first candidate's cultural references (military equipment, stories from a friend who has been working for Red Cross in Afghanistan during the war, Terminator) are not in line with the school's profile, and one of the evaluators asks him: "Are you kind of crazy about war?" (To which he answers: "No, not really.") Moreover, his design techniques, for example a mind-map, are considered simplistic, although the evaluators acknowledge: "It's okay. There are some ideas." On the other hand, the assignment of the second candidate deals with a timely topic (sustainability), and is composed of elements that the school values, particularly several sketches that illustrate the design process. Also, the extra pieces of work, which this applicant shows at the interview session, prove that he knows how to work with design. For example, a piece consists in photos of a model made of acrylic, which has been cut out and scratched to get a frosted look. Although this model is an assignment from the design college, which has been made under the supervision of a teacher, it nevertheless demonstrates that the applicant knows how to make a design model correctly.
As these examples illustrate, cultural capital becomes decisive at the test. Class background and knowledge about dominant culture in society seem to translate into familiarity with the design field. Whereas the applicant who is born into a family with high cultural capital knows the codes of the field and is accepted into the school, the applicant whose social background is not as favorable is unfamiliar with the codes of the field, and thus he becomes excluded. Despite the chubby applicant's ideas and technical skills he is unable to formulate these successfully. In that way, the admission test seems to reproduce social inequality by valuing presentations that requires knowledge about the field, which presupposes cultural capital.

Evaluators' Reactions to the Bourdieusian Approach
When I ask one of the evaluators about the problematic about the importance of cultural capital during a break, he explains: Of course, it's a huge advantage to come from a home of architects with nice things, compared to coming from Ishøj [a suburb south of Copenhagen dominated by public housing] and having been told that "you're good at drawing so you should seek into the design school," and then having a dad who is warehouse keeper. But sometimes, earlier on, we have accepted students who were below level-when one of the applicants that we have first accepted has backed out-and they have turned out to become really good.
As this quote illustrates, the evaluators are perfectly aware of the important part which cultural capital plays at the test. However, the quote moreover points to an awareness of the potential which unlikely students may possess. Compared to students who know the codes of the field beforehand, those who do not may develop more at the school.
When I presented my observations to the evaluators and the administration at the school at an internal seminar we discussed my analysis. According to the evaluators, the Bourdieusian analysis is suspicious of their motives, holding out bleak prospects for whether they can make fair evaluations, and they felt misrepresented by the picture it gives of their actions. If social background, dispositions and habitus always sneak in and bias the valuations made by cultural intermediaries, then how could this be overcome, the evaluators wanted to know. Also, if this apparent bias is what creates not only one's personal but also one's professional foundation for evaluating work, then maybe it is indispensable and should not be attempted to be erased, the evaluators suggested.
According to the evaluators, it is necessary to select a rather homogenous group of students, who have some knowledge about the codes of the field, because it is a prerequisite for the teaching. However, if these codes that appear secret and invisible to outsides could be made visible and accessible, and if this knowledge could be distributed more widely, it would be an advantage for the school because it would give a larger pool of qualified candidates which could raise the standards, the evaluators argued. If some assignments are poor due to lack of knowledge about what is expected it raises the question if knowledge about the expectations of the school is something you can obtain. For that reason, the school has initiated a process of explicating and standardizing its assessment criteria.

S tandardizing Valuations by Means of Devices: Of f icial Cr iter ia and Scor ing Mec hanisms
To further develop the concept of cultural intermediaries, Moor suggests investigating precisely how cultural intermediaries enact their influence. In Moor's words: "there remains a tendency towards vague assertions of a 'shaping' role and reluctance to be more specific" (2012, 570). As an alternative to the Bourdieusian approach, Moor outlines an approach inspired by ANT, which draws attention to the networks that cultural intermediaries form part of. As ANT classically proposes, networks are composed and stabilized as heterogeneous actors become enrolled (Callon 1986). Hence, not only human actors but also non-human agency can be central for defining and upholding a network. For studies of cultural intermediaries, Moor suggests that ANT has potential to expand the scope of empirical investigation considerably, taking into consideration anything that acts as a mediator (Moor 2012, 570). Hence, in contrast to focusing solely on cultural intermediaries (and their habitus), ANT opens up a much wider perspective, where agency is seen as distributed and various types and levels of agency need to be defined empirically.
In the case of the admission test this means that it is not only relevant to study evaluators, their actions and dispositions, but also the institutional set-up which the school provides. Three devices in particular, invented by the school's administration to structure evaluators' assessments, seem relevant to consider: a list of official assessment criteria and two scoring mechanisms based on these criteria. Following Moor, who builds on studies within the 'new new economic sociology,' the two scoring mechanisms may be seen as performative measurement techniques which make some qualities visible while leaving others in the background (Moor 2012, 571; see also Didier 2007). In other words, the scoring mechanisms may be considered as measurement devices as they produce numbers which order applicants that seek into the school. Particularly the second scoring mechanism, which most directly relates to the official criteria, constitutes an attempt to order, standardize and stabilize the assessments that evaluators make.

Official Assessment Criteria
Over the last years, the school has been working to make the admission process transparent by explicating and standardizing its assessment criteria. The school strives to make valuations at the admission test based in a uniform list of criteria to avoid subjective and opaque parameters. As one of the evaluators replies, when an applicant at an interview asks what they think about her work: "We are not going to tell you that. We evaluate you according to these criteria [shows her the list of official criteria]." On the school's website a manual for the admission test can be found, which includes a list of four official assessment criteria. On the webpage it says: We assess your professional talent as a designer and your potential for development. We do this by assessing your abilities to: explore and document. Produce and develop ideas. Treat and develop form, function, materials and digital tools. Disseminate and communicate.
To rank applicants on the foundation of these standardized assessment criteria, the school operates with two scoring mechanisms, one which is used for folders and one for interviews. In that way, the school aims to make evaluators put their qualitative valuations into an ordered system to make their valuations calculable and comparable.

The Scoring Mechanism for Folders
After reviewing folders, evaluators fill out a form for each folder. On this form, the four criteria (see table 1) are restated to draw evaluators' attention to these criteria. Evaluators fill in the date, the number of the applicant, put a cross at either the box 'yes' or the box 'no' as to whether the applicant should be invited to an interview, and give the signatures from both evaluators (see table 2.a). However, in previous years this procedure has resulted in too few interviewees, and the evaluators have had to look through the huge pile of submissions given a 'no' again to find further interview candidates. For that reason the category 'maybe' was invented a few years before this study (see table 2.b). Nevertheless, this category has grown so that it has become a major task to look through this pile of 'maybe' again to find a limited number of extra candidates for the interviews. About two hundred candidates should be invited to the interview sessions, the school has decided. This number is considered small enough to be manageable and large enough to ensure that there is also a selection happening during this second part of the test. Hence, at the first day of the review of folders, a new ranking system is introduced by a secretary from the school administration together with a representative of the evaluators. In 2010, 'maybe' is divided into three sub-categories: 'good maybe,' 'middle maybe' and 1. Ability to explore and document 2. Ability to produce and develop ideas 3. Ability to treat and develop form, function, materials and digital tools 4. Ability to disseminate and communicate 'poor maybe' (see table 2.c). However, these categories cannot be marked on the form (which only contains the boxes 'yes ' and 'no'), and for that reason differently colored post-it-notes are introduced to signal the ranking of the folders. Each form should be supplemented with a colored post-it-note to signal if it is a 'yes' (purple), a 'good maybe' (pink), a 'middle maybe' (yellow), a 'poor maybe' (green) or a 'no' (orange). With this grading system it will be easier to find the two hundred interview candidates.
One of the student observers suggests that the color ranking system should be substituted with numerical ranking, which will be easier to understand. As a compromise, the color ranking system is supplemented with a numerical system: 1 is 'yes' (purple), 2 is 'good maybe' (pink), 3 is 'middle maybe' (yellow), 4 is 'poor maybe' (green) and 5 is 'no' (orange). One of the evaluators objects that with this new grading system it will be the new middle categories that will be used all the time. As the central question concerns the dividing line between 'yes' and 'no,' the sub-categories of 'maybe' only displaces this. In other words, the maybe category accentuates the difficulties of drawing a clear dividing line and can be seen as an attempt to introduce nuance in the process of categorization.
A. The original tick off boxes on the assessment form.

B. Introducing a middle category.
C. Introducing further middle categories.

The Scoring Mechanism for Interviewees
At the interview sessions, another form is to be filled out, which introduces a different scoring mechanism. Again, the four official criteria are stated on the form: 1) ability to explore and document 2) ability to produce and develop ideas 3) ability to treat and develop form, function, material and digital tools and 4) ability to disseminate and communicate. Yet, during this round of the test each criterion is to be given a grade from the 7-step grading scale (-2, 0, 2, 4, 7, 10, 12). Table 3. Illustration of scoring mechanism for ranking interviewees.
Thus, based on their abilities to explore, produce ideas, develop form, and communicate, applicants receive a total score of up to 48, where 8 means that they have passed the test. With this grading system the school can rank candidates on a longer scale (than the scales used for reviews of folders with only 2, 3 or 5 categories) and select the best sequentially. Based on the total score from the interview sessions the school finds its 105 new students. After having graded applicants at the interview sessions, one question recurs among the evaluators. Namely, what the total score of the candidate is: if it is too high or low to possibly entering the school. The examples above with the two candidates who have designed clothing for expat workers and a pavilion in sustainable materials illustrate this point. Based on the number that constituted the demarcation line last year, the evaluators discuss if it is likely for the candidates to become accepted into the school. Thus, despite the scale ranging from 0 to 48 points, the crucial question is where the dividing line will be drawn. This means that evaluators pay more attention to the overall score than to each of the four sub-scores, and sometimes sub-scores are altered to adjust the overall score and thus make candidates' chances to get into the school better (or worse).
In 2009, the dividing line was drawn at 18 and accordingly evaluators used this number as a point of reference the subsequent Grades 1. Ability to explore and document 2. Ability to produce and develop ideas 3. Ability to treat and develop form, function, materials and digital tools

Ability to disseminate and communicate
Total score year. However, in 2010, the lowest graded candidate to be offered a position had a score of 21. On the other hand, the best candidate scored 48. Applicants are not informed about their scores; they are only told whether they have passed-and whether they are offered a position at the school.

Valuations Pr inciples in Practice: Conventions and Evaluators' Parameters
What happens at the specific instances of valuation during the admission test is not a deductive operation of enacting the official criteria, quite the contrary.
As the examples of assessments of two interviewees in the above section on the Bourdieusian approach illustrated, evaluators do not slavishly follow the official criteria. Instead, evaluators discuss the work of candidates and center their discussions on the specifics of the work in case. Sometimes an official criterion is brought into the discussion. Most often, however, it is not until after the assessment has been made that evaluators turn attention to the list of official criteria and the scoring mechanism. While assessing pieces of work, evaluators are either quiet or making comments on the work. In that way, evaluators' bring inductive arguments into the discussion, relating their assessments directly to features of the work in question.
In this section, I present the parameters that evaluators used to make the assessments that I observed. Based on my observation notes, I have made a condensation of evaluators' spoken arguments, grouping these into categories (see table 4). To validate whether these categories formed adequate descriptions of evaluators' valuation principles I presented my overview of evaluators' parameters at an internal meeting at the school. At this meeting evaluators confirmed that these parameters could be seen as the foundation which their assessments were based on. Hence, although evaluators' more specific and tangible principles of valuating are not formalized as the official criteria, they were nevertheless persistent and uniform across the groups of evaluators that I followed, and confirmed by the evaluators.
To describe evaluators' assessment parameters, I use the notion of conventions from Becker (1982). In his book Art Worlds (ibid.), Becker portrays a number of standards within various art forms-such as the number of musicians in symphony orchestras, length of movies, size of canvases in art museums, etc.-and he suggests that these conventions lay the foundation for interactions within the arts. Accordingly, Becker underlines that the division of labor in art worlds is also coordinated by means of conventions. Thus, according to Becker, although art worlds may seem non-formalized and in pursuit of uniqueness, these worlds are characterized by sets of standardized ways of doing things and uniqueness is defined in relation to these shared standards. Hence, conventions are crucial and shared among participants in art worlds. Becker's concept of art worlds resemble Bourdieu's notion of field as it calls attention to the socially defined norms that make up a domain (Becker 2008). However, the concept of art worlds also differs from Bourdieu's approach in important ways. Arguing that art can be seen as work and collective activity, Becker focuses on division of labor, coordination and conventions. On the other hand, Bourdieu views the field as constituted by power struggles over central positions. Hence, Bourdieu's concept offers a critical approach while Becker's concept outlines an empirical and descriptive approach. Accordingly, Becker's concept of art worlds has been criticized by Bourdieu who characterized it as "pure descriptive, even enumerative" (Bourdieu 1996, 205). Thus, whereas Bourdieu's critical perspective can be used to portray evaluators' actions as problematic, as has been illustrated above, Becker's interactionist view consists in describing art worlds empirically. In the case of the admission test this means that the study becomes a study of critique rather than a critical study.
Becker's interactionism is in line with pragmatism, not only because the two traditions share philosophical roots but also because their ethnographic methodologies overlap and they share an interest in studying how people do things and thus how values and conventions become installed and maintained. By giving Becker a twist in a sociomaterial direction (Hennion 1997), his approach to studying art as collective action may fall under the category of pragmatism (Heinich 2014). Accordingly, Becker's proposal for studying art as work may resonate with the pragmatist idea of seeing valuation as an action (Muniesa 2012), and his focus on conventions may be one way of tracing the contingencies of values (Herrnstein Smith 1988). To follow this route, let us look at the assessment parameters which evaluators at the design school operate with (table 4).

Materials of the Packaging
First of all, materials of the packaging of the folder are important: Digging in the pile, the first evaluator states: "Now comes the envelopes. Five envelopes. This will be fast." Pointing at the envelopes, the lecturer explains to me: "It means something. Size matters. They haven't used A3 although they have the possibility to do so. [The five folders are made on A4 printer paper packed in office envelopes]. That shows how important you think it is. That means you ask to be excluded." Flicking through the pages, the lecturers access the first envelope-folder in two minutes, categorizing it as a clear "no." Nowhere in the instructions for applicants does it say that the packaging of the submission is crucial, but evidently it is. As heaps of submissions are reviewed it becomes obvious that large folders in cardboard, filled with thick A3 pages, produce much higher expectations than thin A4 envelopes on printer paper. The importance of packaging and paper is considered too difficult, or unnecessary, to put into words.

Basic Skills
Secondly, basic skills, particularly drawing skills, are essential: Pointing at a drawing, the first evaluator comments: "I think everything has been copied." The second evaluator agrees: "Yes, here, for example, the feet have not been included in the drawing." [Some square-looking sneakers serve as feet in the drawing]. The first evaluator continues: "It's only things that have been copied using tracing paper, right." Nuancing their negative assessment, the second evaluator suggests: "Yes, but there's a certain compositional ability and it's staged delicately, argument, the second evaluator suggests: "Sampling is one of the ways to move forward." Looking at the drawings the first evaluator is not convinced, but outlines a compromise: "You can have a 4." Reluctantly, the second evaluator accepts: "Then you'll have to return half of the sourdough." While recording the score, the first evaluator comments: "It's nothing personal." Still, the second evaluator rounds off: "I need a moment to collect myself." As the discussion in this example illustrates, it is difficult to convince evaluators of other qualities in the work if it lacks basic skills. In other words, basic skills are hard to compensate for, and drawing abilities are seen as an indication of basic skills. Drawing abilities constitute a starting point, which the first evaluator in this example is unwilling to look behind. If applicants demonstrate a classical drawing style it is considered a good starting point. If, on the other hand, applicants have drawn clumsily it is seen as a clear sign of weakness as this example illustrates. However, it is also considered a weakness if the drawing skills are too schooled; if an acquired style stands in the way of the applicant's own expression.

Several Approaches
In continuation of basic skills, working with several approaches to design is valued: Puzzled by a submission, the first evaluator asks: "If I buy it then what can I spray? My armpits? My bicycle chain? It's an imaginary product for a campaign." The second evaluator moderates: "Yes, the basic idea is not exactly obvious, but it gets through all the disciplines." As the example illustrates, regardless of the idea behind the design, mastering a breadth of disciplines; demonstrating different techniques, is seen as a good sign. One-dimensional assignments are criticized exactly for their one-dimensionality, even though they may excel in the one discipline which they embody. If the entire folder is created in the same style, or if the works in the folder are too similar, it is valued negatively. Also, if applicants work digitally it is crucial to use the computer as a tool, instead of using computer programs to standardize works.
To illustrate how crucial skills and abilities in a range of disciplines are, the interview with the applicant who achieves the highest score at the test may serve as an example: Before the candidate comes in, the evaluators scroll through the assignment on the computer. The assignment is about refugee tents. "On the face of it this looks quite promising. At least it's someone who can draw" the first evaluator states. The candidate comes in. He is a guy in his mid-twenties in Converse shoes. Immediately, when he begins introducing the assignment, the evaluators start asking questions about the technicalities of his work: are the drawings based on photos, are they marker drawings, which drawings are made digitally. Asked about his background, the candidate describes attending a drawing course many years ago and having worked abroad in the computer game industry for several years. "Can you work in 3D?" one of the evaluators asks. "Yes, you name it," the applicant replies. After the interview, the first evaluator affirms: "He would make a fine student." The second evaluator continues: "I was thinking if he has too much energy." The first evaluator takes over: "He draws brilliantly." "10-12?" the second evaluator asks [see table 3]. "Yes, clearly," the first evaluator replies. The applicant achieves the highest possible score: 48 points.
What is exemplified in this case is that drawing abilities and abilities in diverse approaches are highly valued at the test. Basic skills together with mastering of several disciplines make the candidate in question strong and impossible to disagree on. In this case, the evaluators' only doubt is whether the candidate "has too much energy" for the school. His qualifications are absolutely adequate.

Visualization
A further tangible parameter is visualization, in contrast to verbalization: Reviewing a folder, the first evaluator states disapprovingly: "The process is in words." The second evaluator agrees: "If you want to make something like this it has to be explained graphically with arrows and such." First of all, the basic idea in the submission should be clear and relevant. Preferably, the idea should concern an actual problem and if it is related to a user group it is considered a plus. Conversely, if the evaluators do not get the point of the submission, if there is no problem to be solved or if a problem is invented, then it counts against. In the above example, the candidate designs an alternative scarf in relation to the debate over the hijab. By doing so, the candidate locates a problem, which is considered relevant because it is topical. As this exemplifies, ideas are often connected to a zeitgeist. On the last day of reviewing folders, an evaluator comments: "Soon we'll have an assignment with stressed homeless people in burquas." Yet, it is not enough to locate an actual problem. As the above example illustrates, candidates should also be able to design something that treats the identified problem. In the above case, the intention to make an alternative headscarf turns into a design of something else, namely a hood. In that way the idea is not developed properly. Throughout the test, development of the design idea constitutes one of the most commonly raised objections to candidates' work. Particularly, submissions should contain several ideas, which show alternatives to the idea that is in focus. But often, candidates do not live up to this criterion of outlining multiple ideas: Reviewing a submission that consists in a battery for iPhones one of the evaluators announces: "There's not enough in it . . . There is an idea alright, but it's not developed very much.
[Reads from the list of official criteria] 'Produce and develop ideas.' When you don't have your charger [pauses], but what about other alternatives, more ideas, for example a small dynamo on the bike or solar cells?"

Process and Development
As the above example with the iPhone battery illustrates, ideas should be multiple and thus demonstrate development. In other words, locating a problem and coming up with an idea should be followed by an outline of a process. Accounting for the design process is a feature of assignments that is highly valued by the evaluators: Unpacking a submission, the first evaluator bursts out: "Look at this! This is our result!" [Laughs] The second evaluator responds: "You are kidding me! God damn it! Reverse! Reverse the process! It should have proceeded from here to here" [pointing from the last to the first picture]. "She has a fantastic process and then she kills her project and makes a product for IKEA," the first evaluator agrees.
Process and development are essential for the evaluators. Applicants should experiment and investigate, slowly approaching the subject. Ideas should be worked out and the steps in the process should be demonstrated. To show the gradual development and intermediate steps, submissions should contain sketches, preferably originals. Applicants may fall into two pitfalls when they do not include sketches, either making submissions with no development or making submissions with sudden jumps or giant leaps. Both lack of sketches and sudden jumps are considered to be severe weaknesses. Ideally, assignments should include a detailed, varied and rich account of considerations and steps during the entire process of developing a design.

Rigour, Ingenious Solutions, Aesthetic Sense, Potential and Personality
Besides these principles of looking for materials of the packaging, basic skills, a scope of disciplines, visualization, ideas, process and development, the evaluators also value rigour, ingenious solutions and aesthetic sense. Last but not least, evaluators seek potential and personality. Potential means that the submission should show raw material and indicate being made by a searching, teachable person. Personality means that the submission should give an impression of the person behind it. On the contrary, if a submission is impersonal, or if it is too absorbed in personal emotions, it is valued negatively.

Concluding Remar ks
To address the evaluations made at the admission test at the design school, the paper has outlined a pragmatist position as its theoretical foundation. Pragmatism not only constitutes a common ground between economic sociology and cultural sociology, it moreover offers a way of addressing values as contingent. Thus, a starting point for this paper has been Barbara Herrnstein Smith's notion of contingencies of values. In opposition to seeing aesthetic value as subjective, Herrnstein Smith suggests that values are contingent: "contingent in the sense that it is a function . . . of the states of numerous particular systems interacting at a particular time and place" (1988, 183, emphasis in the original). Subscribing to this view, the paper has suggested investigating the admission test as an exercise in contingencies of values where official devices and professional conventions meet in the makings of assessments.
Moreover, the paper has employed theoretical inspiration from the literature on cultural intermediaries to describe the role of evaluators at the test. Whereas the concept of cultural intermediaries originates from Bourdieu, a new strand within this line of research suggests turning attention to ANT in the form of the 'new new economic sociology' and thus pragmatism. Adding to this body of literature, the paper has demonstrated how evaluators at the admission test make assessments that are based in other variables than their own habitus. That is, rather than suggesting that the evaluators make subjectively biased assessments, as the critical Bourdieusian approach proposes, this paper has shown that evaluators make assessments based on official devices, conventions of the design world and impressions of candidates and their works.
Yet, as the first part of the analysis has illustrated, it is indeed possible to apply a traditional Bourdieusian framework to evaluations at the admission test, which makes visible a form of social stratification at the test. With this perspective, applicants are seen to be selected because of their cultural capital, or excluded because of their lack of cultural capital. Whereas this story is fascinating because of its critical edge, it makes the test seem highly problematical, and the evaluators do not find this story do them justice. Moreover, whereas the Bourdieusian perspective is sensitive to power struggles and reproduction of cultural preferences, this perspective does not take into account that the school actively strives to make its assessments standardized and non-subjective.
Hence, the second part of analysis has looked into the school's attempt to overcome what is perceived as subjective bias in aesthetic valuations by introducing a standardized set of assessment criteria and ranking systems. With four official criteria, the school explicitly aims to make its evaluations formal, uniform and ideally objective. Turning attention to these criteria and the two ranking systems that they are accompanied by, this part of the analysis has considered the quantification of qualitative, aesthetic features. Informed by the 'new new economic sociology' as it has been taken up in relation to cultural intermediaries this part of the analysis has focused on the performative effects of devices.
In the third part, the analysis has shown that in the actual practices of valuating applicants' work, evaluators do not employ the devices that the school has introduced. Rather, evaluators rely on a set of conventions based in their professional expertise, which the analysis describes with reference to Becker's work on conventions within art worlds. Although these valuation principles are not as formalized as the official criteria, they are nevertheless uniform across the group of evaluators, and they make good sense in practice. Evaluators at the test do not employ the devices of the official criteria and the scoring mechanisms which are used to rank applicants until after assessments have been made. When making their assessments evaluators rely on a set of conventions which allows them to address the specifics of the unique objects they encounter. Once assessments have been made and a grade is to be given evaluators turn attention to the official criteria and the related scoring mechanisms.
As the analysis shows, evaluations made at the test are not a simple matter of overcoming subjective bias by means of objective criteria and measurement devices. Rather, evaluations are contingent as they compose the functions of multiple variables. That is, the pragmatist approach does not accept the common storyline about the test: that evaluators' subjective assessments biased by cultural capital, habitus and personal preferences can be substituted with objective assessments by implementing standardized assessment criteria and devises for scoring candidates in accordance with these criteria. Instead, the pragmatist approach to seeing values as contingent suggests that these two competing logics co-exist at the test, together with several other evaluative principles. Hence, the two competing logics of standardizing assessments to avoid what is believed to be subjective bias could be described in terms of the multiple orders of worth literature as an industrial logic of measuring professional capabilities and an i n s p i r a t i o n a l l o g i c o f v a l u i n g t h e u n i q u e t h a t " e l u d e s measure" (Boltanski and Thévenot 2006, 159; see also Stark 2008). In the words of Herrnstein Smith the co-existence of multiple orders of worth is exactly what contingencies of values are about, as she describes this as "numerous particular systems interacting at a particular time and place" (1988,183).
In the case of the admission test, the analysis has proposed seeing evaluators' assessments as based in and maintaining conventions within the world of design, rather than a reflection of evaluators' own habitus. As the third part of the analysis has shown, evaluators' assessment parameters are numerous, uniform across the group of evaluators and employed when relevant. Hence, evaluators' parameters present a whole set of conventions whereas the school's official criteria are differentiated and thus made comparable (Didier 2010). This means that the translation of assessments into a grading based in this formal apparatus risk taking the focus away from what evaluators are actually looking at when assessing the work of candidates. While evaluators are guided by a wide repertoire of design conventions when assessing unique, singular products, the devices that the school introduces becomes effective in expressing a certain characterization of what they describe (Didier 2007).
Thus, values as expressed in evaluations at the admission test are contingent because they employ standardized assessment criteria and measurement devices while being founded on conventions within the world of design and initiated by the specifics of the objects that are being evaluated. In that way, evaluations can be described as interactions not only between various contingent background factors, but also between these factors and the pieces of work which are being tested. By testing the effects of the works, in relation to multiple factors of conventions and devices, contingencies of value at the admission test is not a social game but a game of matching works with ever contingent aesthetic values.